Stick-Breaking Variational Autoencoders

Nalisnick, Eric; Smyth, Padhraic

doi:10.48550/arxiv.1605.06197

Cited by 16 publications

(20 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For NP and ANP trained on functions generated from GP, we illustrate the weight norm of the decoding layer right behind the latent variables in Figure 5. The sparsely-coded decoder implies the redundancy of the stochastic path due to the component collapsing behavior referred to in [40,24]. This phenomenon can be explained by the information preference problem [7,73] where the information flow is concentrated on the deterministic path with the tendency to ignore the stochastic path.…”

Section: Encoder-decoder Pipelinementioning

confidence: 99%

Meta-learning Amidst Heterogeneity and Ambiguity

Go¹,

Yun²

2021

Preprint

View full text Add to dashboard Cite

Meta-learning aims to learn a model that can handle multiple tasks generated from an unknown but shared distribution. However, typical meta-learning algorithms have assumed the tasks to be similar such that a single meta-learner is sufficient to aggregate the variations in all aspects. In addition, there has been less consideration on uncertainty when limited information is given as context. In this paper, we devise a novel meta-learning framework, called Meta-learning Amidst Heterogeneity and Ambiguity (MAHA), that outperforms previous works in terms of prediction based on its ability on task identification. By extensively conducting several experiments in regression and classification, we demonstrate the validity of our model, which turns out to be robust to both task heterogeneity and ambiguity.

show abstract

Section: Encoder-decoder Pipelinementioning

confidence: 99%

Meta-learning Amidst Heterogeneity and Ambiguity

Go¹,

Yun²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…As for the prior, the standard VAE takes the normal distribution as the prior, which may lead to the phenomena of over-regularization and posterior collapse and further affect the performance for density estimation [5,11]. In the early stage, VAEs apply more complex priors, such as the Dirichlet process prior [18], the Chinese Restaurant Process prior [19], to improve the capacity of the variational posterior. However, these methods can only be trained with specific tricks and learning methods.…”

Section: Bayesian Pseudocoresets Exemplar Vaementioning

confidence: 99%

ByPE-VAE: Bayesian Pseudocoresets Exemplar VAE

Wen³

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent studies show that advanced priors play a major role in deep generative models. Exemplar VAE, as a variant of VAE with an exemplar-based prior, has achieved impressive results. However, due to the nature of model design, an exemplar-based model usually requires vast amounts of data to participate in training, which leads to huge computational complexity. To address this issue, we propose Bayesian Pseudocoresets Exemplar VAE (ByPE-VAE), a new variant of VAE with a prior based on Bayesian pseudocoreset. The proposed prior is conditioned on a smallscale pseudocoreset rather than the whole dataset for reducing the computational cost and avoiding overfitting. Simultaneously, we obtain the optimal pseudocoreset via a stochastic optimization algorithm during VAE training aiming to minimize the Kullback-Leibler divergence between the prior based on the pseudocoreset and that based on the whole dataset. Experimental results show that ByPE-VAE can achieve competitive improvements over the state-of-the-art VAEs in the tasks of density estimation, representation learning, and generative data augmentation. Particularly, on a basic VAE architecture, ByPE-VAE is up to 3 times faster than Exemplar VAE while almost holding the performance. Code is available at our supplementary materials.

show abstract

“…The simple CDF of Kuma also makes the reparameterization trick easily applicble [51,79,62]. Lastly, KL-divergence between Kuma and Beta distribution can be approximated in closed form [70]. We fix βt = 1 to ease optimization since the Kuma and Beta distributions coincide when…”

Section: Recurrent Variational Model With Differentiable Binary Laten...mentioning

confidence: 99%

“…The nodes, termed pivotal states, are the most critical states in recovering action trajectories [17,47,32]. In particular, given a set of trajectories, we optimize a fully differentiable recurrent variational auto-encoder [19,34,51] with binary latent variables [70]. Each binary latent variable is designated to a state and the prior distribution learned conditioning on that state indicates whether it belongs to the set of pivotal states.…”

Section: Introductionmentioning

confidence: 99%

Learning World Graphs to Accelerate Hierarchical Reinforcement Learning

Shang,

Trott,

Zheng

et al. 2019

Preprint

View full text Add to dashboard Cite

In many real-world scenarios, an autonomous agent often encounters various tasks within a single complex environment. We propose to build a graph abstraction over the environment structure to accelerate the learning of these tasks. Here, nodes are important points of interest (pivotal states) and edges represent feasible traversals between them. Our approach has two stages. First, we jointly train a latent pivotal state model and a curiosity-driven goal-conditioned policy in a task-agnostic manner. Second, provided with the information from the world graph, a high-level Manager quickly finds solution to new tasks and expresses subgoals in reference to pivotal states to a low-level Worker. The Worker can then also leverage the graph to easily traverse to the pivotal states of interest, even across long distance, and explore non-locally. We perform a thorough ablation study to evaluate our approach on a suite of challenging maze tasks, demonstrating significant advantages from the proposed framework over baselines that lack world graph knowledge in terms of performance and efficiency.

show abstract

Stick-Breaking Variational Autoencoders

Cited by 16 publications

References 0 publications

Meta-learning Amidst Heterogeneity and Ambiguity

Meta-learning Amidst Heterogeneity and Ambiguity

ByPE-VAE: Bayesian Pseudocoresets Exemplar VAE

Learning World Graphs to Accelerate Hierarchical Reinforcement Learning

Contact Info

Product

Resources

About